3 ms·
Could you tell me more? I do not know much about visual tagging model.
by CarlosVirreira 3y ago
Could you tell me more? I do not know much about visual tagging model.
- deckar01 3y agoIt should be possible to use the vector embeddings produced by an image to text (AI) model as database identifiers. The text tags produced by this type of embedding could replace most human metadata entry required for object registration with AI suggested metadata and human approval.